Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Kimi-K3 on HuggingFace

huggingface.co

Post reply on HN
Print view View original

551–560 of 588 posts

Re: Kimi-K3 on HuggingFace

#551

Post by nullbio »

ataoz wrote: post #530

I asked "Tell me about yourself" on HF. This is the response... Curious. > Kimi K3: I'm Claude, an AI assistant created by Anthropic. I'm built to be helpful with a wide range of tasks—things like writing and editing, answering questions, coding, analysis, brainstorming, explaining concepts, math, and creative projects.

This really means nothing.

You can ask the same thing in Chinese to Opus and it will tell you it's DeepSeek, for example.

Things leak into the training data and models hallucinate.

Re: Kimi-K3 on HuggingFace

#552

Post by wolttam »

docheinestages wrote: post #67

Given the frontier-level capabilities of Kimi K3, I'm wondering if it's possible to extract the core capabilities (fundamental reasoning and tool calling) of the model into a smaller one that consumer devices could run? Not sure exactly how, but either by heavy distillation or some other surgical method since Kimi has a Mixture of Experts architecture. I think it's very valuable to have a smaller model that doesn't h…

> smaller model that doesn't have any domain knowledge or facts built into its weights

I’m not sure it works this way. Language modelling itself is a “domain”, and if it didn’t have a grasp language it wouldn’t be able to do anything else.

I also think a “reasoning engine” that had to reason through everything from first principals would likely be extremely inefficient.

It’s good went models have domain knowledge and expertise - and all of their reasoning flows downstream of that

Re: Kimi-K3 on HuggingFace

#553

Post by sebmellen »

cobbzilla wrote: post #250
sebmellen wrote: post #219

Earlier quoted context omitted.

Gotcha. But to be clear, you’re talking only about energy usage, correct?

Yes, what other opex is there? It will have good ventilation, I’m not worried about cooling.

I thought you were talking about renting a dedicated server from a hosting provider, not racking it yourself. My bad!

Re: Kimi-K3 on HuggingFace

#554

Post by Recoordinates »

The release itself is great news, but I'm even more interested in the ecosystem that forms around it. Open weight models tend to improve much faster once people start building inference stacks, quantizations, evaluation suites, and fine tunes. It'll be interesting to see where Kimi-K3 stands in six months compared to today's benchmarks.

Re: Kimi-K3 on HuggingFace

#556

Post by dannyw »

mock-possum wrote: post #392
gorgmah wrote: post #390

Earlier quoted context omitted.

good find! This sounds a bit like what Meta was doing with the earlier Llama models? There is also this paragraph in their licence that is smart marketing-wise: > 3. If the Software (or any derivative works thereof) is used for any of the Licensee's commercial products or services that have more than 100 million monthly active users, or more than 20 million US dollars (or equivalent in other currencies) in monthly re…

Is that even enforceable?

What's wrong with it? Licenses like the MIT license already requires you to attribute, the only difference is the "prominent" part.

Re: Kimi-K3 on HuggingFace

#557

Post by nektro »

ttoinou wrote: post #142
bjackman wrote: post #104

Earlier quoted context omitted.

I agree but worth noting that it's never gonna be very practical to run LLMs like this at home. Unless we have some sort of design breakthrough, the only "sensible" way to run them is at high batch levels on shared HW. Like, yeah if I could spend a few grand on such a GPU I probably would coz I'm a rich nerd, but I'd acknowledge it as an extremely inefficient luxury, kinda like a sports car. So I think you could say…

We could make LLM inference 100x cheaper to run at home efficiently, but that solution might need to be updated every 1-2 years, whereas current GPU are useful for various others tasks and last longer

> We could make LLM inference 100x cheaper to run at home efficiently

do you genuinely think that's going to happen?

Re: Kimi-K3 on HuggingFace

#558

Post by dannyw »

veber-alex wrote: post #501
fnordpiglet wrote: post #445

Earlier quoted context omitted.

I have to say cc opus 5 is abysmal. It talks to itself incessantly, gets stuck in minutia, fails to understand problems clearly and makes steering mistakes constantly. It also has a weird behavior where it says “ok I know exactly what to do and I will start now,” then sits waiting for user input. If you’re not on the ball you’re constantly losing 5m/1h cache. Just give me back 4.6.

I still see Opus 4.6 available in CC. Also I have been using Opus 5 for the last few days and I find it work fine. It's a little verbose but the code quality is good.

You can even use /model to use Opus 4.5 if you want. Broadly though I find Opus 5 and most of their point releases (4.7 being the exception) to be excellent and upgrades.

Re: Kimi-K3 on HuggingFace

#559

Post by dannyw »

great_psy wrote: post #399
horsawlarway wrote: post #295

Earlier quoted context omitted.

I don't think the goal of this data is necessarily model improvement. I think it's marketing, advertising, and product refinement. Ex: all the things Google wants your search data for. It's somewhat silly to think the value of that data has changed much. Advertisers want to know what's popular and getting clicks and attention. Competitors want to know what features are getting used in their markets. In the simplest c…

I wonder how much less useful it is if I use those models for open code or similar. What are you really learning about me, other than the fact that I am a technical person, which you could know by the fact that I signed up for open router to start with.

Do you run your coding harness in a completely segregated sandbox? Even folder names of what projects you have, what projects you actually work on, etc are very useful for targeted advertising.

Re: Kimi-K3 on HuggingFace

#560

Post by nextaccountic »

mdasen wrote: post #309
walrus01 wrote: post #64

Earlier quoted context omitted.

There are a number of use cases where sending the contents of your context and prompts (and the resulting output) to a 3rd party service is off the table as an option, and people will compromise speed for data sovereignty. And not everyone's electricity is equally expensive, I pay about $0.075 USD per kWh. It would for example cost me about $48 a month of electricity (not counting cost of cooling) to run a quad socke…

That's an unusually low electric rate for the US - way below the lowest state average which is Idaho at 12.4 cents. It's certainly possible that you are getting 7.5 cents including delivery, but I've had friends say that they're "getting 13 cents per kWh" here in Massachusetts, but that's just the supply rate and the delivery is another ~18 cents. There are parts of states like Grant County Washington that have cheap…

> That's an unusually low electric rate for the US

What about buying your own solar panels and batteries?

Post reply on HN